Multi-omics reveal SHBG links to subclinical atherosclerosis in men with HIV

In a finding that could reshape how scientists think about heart disease risk in people living with HIV, a large multi-omics study has revealed that higher blood levels of sex hormone-binding globulin, a protein long dismissed as little more than a carrier molecule for testosterone and estrogen, are strongly linked to less atherosclerosis in the arteries of middle-aged and older men. The research, published in Genome Medicine, went far beyond a simple hormone measurement: by weaving together gut microbiome sequencing, nearly a thousand circulating metabolites, and almost three thousand plasma proteins, the team uncovered an entire molecular ecosystem that travels with this hormone-binding protein and appears to protect the arteries.

The study drew on 321 men from the MACS/WIHS Combined Cohort Study, one of the longest-running observational research programs on HIV in the United States. Roughly 65 percent of the participants were living with HIV, and the group’s median age was 62 years. Using high-resolution B-mode ultrasound, the investigators examined each participant’s carotid arteries, the large vessels that carry blood to the brain, and looked for plaque buildup with an arterial wall thickness exceeding 1.5 millimeters, the standard marker of subclinical atherosclerosis, meaning artery disease that has not yet caused symptoms. Just under a third of the men, 31.5 percent, had detectable carotid plaque.

What makes the study methodologically striking is its layered design. The researchers first measured 14 serum sex hormones along with sex hormone-binding globulin, or SHBG, the glycoprotein that binds sex steroids in the bloodstream and regulates how much free, biologically active hormone circulates in the body. Then, in a subset of 312 men, they profiled 986 plasma metabolites using liquid chromatography-tandem mass spectrometry, a technique that separates and identifies small molecules with high precision, and quantified 2,883 plasma proteins on the Olink Explore 3072 platform, an affinity-based proteomics assay that measures proteins across inflammatory, metabolic, and cardiovascular pathways. Stool samples underwent metagenomic sequencing, allowing the team to identify gut microbial species and their relative abundances.

The central question was whether sex hormones relate to artery plaque differently in men with HIV compared with men without it, and whether the molecular fingerprints of those hormones might explain the connection. The team built what they call omics scores: linear combinations of the gut microbial species, metabolites, and proteins most strongly associated with a given hormone. If a hormone’s downstream molecular signature is also linked to plaque, that signature offers clues about mechanism, not just correlation.

The headline result concerned SHBG. In men with HIV, each one-standard-deviation increase in SHBG was associated with 40 percent lower odds of carotid plaque, with an odds ratio of 0.60 and a 95 percent confidence interval of 0.41 to 0.90. No such protective association appeared in the men without HIV, and the pattern of associations for the other sex hormones diverged between the two groups as well, hinting that HIV itself, or its long-term interplay with antiretroviral therapy and chronic immune activation, alters how the endocrine system relates to vascular health.

The multi-omics layer then revealed what SHBG is traveling with. Higher SHBG levels tracked with a measurably different overall gut microbial composition, including lower abundance of species from the genera Prevotella, Fibrobacter, and Coprococcus. They also tracked with higher levels of certain circulating metabolites, predominantly lipids and carnitine-related compounds, molecules that sit at the intersection of fat transport and mitochondrial energy metabolism, and with a protein profile enriched in the cell-cell adhesion pathway, the molecular machinery that governs how immune cells stick to blood vessel walls, a key early step in plaque formation.

Crucially, several of the individual molecules associated with SHBG were themselves associated with plaque in men with HIV. The microbial species Mediterranea massiliensis, phosphatidylcholine-based lipids, and proteins involved in immune response pathways all appeared in both lists, connecting SHBG to artery disease through three independent biological domains. When the researchers consolidated this signal into the three omics scores, the results converged: all three scores were inter-correlated with one another and each was inversely associated with carotid plaque in men with HIV. A species score, a metabolite score, and a protein score, built from completely different measurement technologies, all pointing the same direction.

The contrast group told a different story. Among men without HIV, the only significant hormone-plaque link was estrone-sulfate, a sulfated form of estrogen, which was positively associated with plaque, with an odds ratio of 3.80 and a 95 percent confidence interval of 1.41 to 10.22. Notably, estrone-sulfate showed no associations with any gut microbial species, metabolites, or proteins, suggesting its relationship to artery disease may run through a different mechanism, or that the study lacked the statistical power to detect one. This asymmetry between the two groups is one of the paper’s most provocative implications: the cardiovascular meaning of a given hormone appears to depend on HIV status.

The findings carry real-world weight because people living with HIV face an elevated burden of cardiovascular disease that traditional risk calculators do not fully capture. Even with viral suppression maintained by antiretroviral therapy, chronic inflammation, immune dysregulation, and metabolic changes persist, and heart attacks and strokes occur more often than expected. If SHBG, a molecule that can be measured cheaply in serum, genuinely marks, or perhaps mediates, a protective vascular state in this population, it could become part of risk stratification and eventually a therapeutic target.

The word “perhaps” matters here. This is a cross-sectional observational study: hormones and omics profiles and plaque were measured at the same time, so the data cannot prove that high SHBG causes less plaque. Reverse causation is plausible, since systemic illness, inflammation, and metabolic dysfunction are known to lower SHBG levels, meaning that plaque itself could conceivably drag the protein down. The authors’ exploratory mediation analyses, testing whether the omics scores statistically explain the SHBG-plaque association, are suggestive rather than definitive. Longitudinal follow-up, and ideally interventions that raise SHBG, would be needed to establish causality.

Still, the study exemplifies where cardiovascular research is heading. Rather than testing a single biomarker against a single outcome, the multi-omics approach maps the shadow a molecule casts across the gut microbiome, the metabolome, and the proteome, then checks whether those shadows fall on disease itself. SHBG, often treated as a passive transport protein, emerges from this analysis as a hub connected to gut ecology, lipid metabolism, carnitine handling, and vascular immune biology. Whether the protective signal holds over time, and whether it extends to women with HIV and other populations, will be the natural next questions. For now, the study offers men living with HIV a new molecular clue to a persistent clinical puzzle: why their arteries age faster than their calendar risk factors predict.